This project has been suffering from a complete lack of effort and resources on my part. Which has basically resulted in little obvious progress for the last couple of years. Which is not ideal.
There have been many factors that have caused this and if you would like to read on, then please do so below:
First impact was a big hit and a complete reduction of enthusiasm: was that I was unable to secure any funding to continue development. I had a few meetings with The Business Gateway here in Orkney and they were very helpful indeed. And I can’t fault them on that front. But really the crux of the problem to gaining more funding, was that I did not really have anything to invest upon. I needed to have some tests systems installed and results to show the gains. So the project needed to produce some units and get them installed in the field for a year or so. Then we could re-investigate funding.
Next came issues with the hardware. So in many respects, not getting the development funding was good. Otherwise I would have been in trouble. The hardware I had chosen to control the valve, microprocessor wise, was not really up to the task, and so I needed to look for an alternative. This was in the form the the Espressif ESP8266 module. Mostly because it has WiFi built into the silicon, making it an absolute ideal choice for providing logging information out and Over-The-Air (OTA) software updates and configuration. However, I had issues with a steep learning curve and major issues with software libraries for this device.
Although the ESP8266 can be coded within the ubiquitous Arduino IDE and use many of the libraries, there are issues. The ESP8266 has an even driven core, which the Arduino UNO (target of many Arduino libraries) does not have. This prevents many operational libraries from functioning and typically causing processor exceptions or the WDT timer to trigger and cause a processor reset. I also needed a scheduling library to simplify system operations within the ESP. I did not with to learn FreeRTOS in addition to trying to learn a new processor, and so became stuck, until my knowledge could catch up. The ESP was the ideal choice, but there were huge issues with scheduling my code operations.
I discovered a very good scheduler about a year back called Esparto (now no longer developed), by a great programmer Phil Bowles, and began developing my code based on this very capable library. And things progressed slowly. Finding time outside of work and family issues, made development much slower than expected.
I also need to learn about MQTT, as this was the ideal network protocol for data logging and parameter control. So that took a little while to learn and progress. This forced a ‘spin off’ project to install full MQTT and data logging facilities (via MariaDB SQL). This has taken time to implement the relevant hardware (Raspberry Pi). And I needed to implement various little sensor modules (also based on the ESP8266) to test my MQTT platform and H4 scheduler. Using H4 software Phil & I quickly discovered that all the MQTT libraries for the ESP did not work correctly to the MQTT specifications, or were just darn right rubbish. I’m not going to name names, but out of around 6 main stream ESP MQTT libraries, none were really usable. This resulted in Phil producing a new project to fix the ESP MQTT issues and this resulted in the PangolinMQTT project and library.
I am now at a stage where I’m happy with both the ESP hardware as a platform and all the required ESP libraries to produce functional and stable code.
The next stage will be to produce a prototype triac control board for the motorised valve and link it all together. The next stage would be to modify one (or more) of my heating systems and install the valve to fully debug and prove the software and overall operation.
On my house, I have a Nibe GSHP which would benefit hugely from this proportional valve. However, the company that installed it (Black Isles Renewables), used crimp pipe connections for the installation. And although this makes for a speedy installation and reliable pipe connections, it makes it impossible to modify the installation. This would require getting the original company back in again at huge cost, or invest in the same tooling. Again a large cost. It is really very annoying that the installation company used these crimp connections. Although they would probably tell me for reliability and cost reasons (only a saving and extra profit from their point of view), it makes repairs and change impossible without major reworking of the original installation.
As I am slowly refurbishing this property, I have held back on replacing this original crimped pipework, with more traditional push fit and sweated connections, until I actually refurbish this section of the property. However, I have seen a 30% energy reduction on the heat pump energy usages by replacing all the 22mm core pipework with 28mm pipework. And this extends right back until the connections with the crimped pipework.
But if I want to install the proportional valve any time soon, I will need to bite the bullet and remove the crimped pipework for traditional pipework.
And I’m still looking for development investment and volunteers to progress this project in a more timely fashion. Funding the project via profit from my business slows the process down immensely, especially with the recent issues of Covid-19 affecting my access to work and supply of general work materials. This affects my profit margins and available time. Lock downs have given me plenty of spare time, but no money to invest in the required parts. And family issues continue to consume lots of my spare time, but I can see an end to issue, probably around Q2 of 2021, presuming Covid-19 does not get in the way any further…..